avatar

dcollins

Platinum Blonde

Posts: 2,326 Member Since: 27/01/2011

Lead

Apr 6 12 5:51 PM

Tags : :

Well, after the excellent info on 'normal' soldering, I think it's time for users to weigh-in on their approaches and tools for surface-mount rework.

I have recently entered into this world and can say for sure that it's annoying.  I know with time that I will become more comfortable with it, but at the present time I dread having to go in there and fix stuff.

Present needs range from 0402's to 144 pin LQFP.

I note with no small irony that right as my eyes began to go, electronic components became nearly invisible.

The upside is that if you drop a part, there is no point looking for it on the floor...............


DC

Quote    Reply   
avatar

sinder

Silverado

Posts: 80 Member Since:04/02/2011

#1 [url]

Apr 6 12 8:31 PM

SMD = PITA.  Been in pub, so no help now.  Good topic.

When I say pub, I mean a real one, not the PRW virtual hangout.

Ian Tea drinker, homebakes preferred

Quote    Reply   
avatar

andy peters

Silverado

Posts: 54 Member Since:07/02/2011

#2 [url]

Apr 6 12 10:10 PM




Well, after the excellent info on 'normal' soldering, I think it's time for users to weigh-in on their approaches and tools for surface-mount rework.I have recently entered into this world and can say for sure that it's annoying.  I know with time that I will become more comfortable with it, but at the present time I dread having to go in there and fix stuff.Present needs range from 0402's to 144 pin LQFP.I note with no small irony that right as my eyes began to go, electronic components became nearly invisible.The upside is that if you drop a part, there is no point looking for it on the floor...............

-dcollins

Neither threat nor menace. Think of it as an opportunity. It is also the way of the world.

Tools you need, in no particular order (because you need them all):

a) a good stereoscope. You can't solder what you can't see.
b) Metcal/OKI soldering tool with a range of tips. You're in the big leagues. You need tools. You don't skimp on microphones and you pay $$$ for fancy cable, so buy a real soldering station.
c) solder wick.
d) flux. You want flux. Flux is your friend. 
e) bite the bullet and get the lead-free solder,as the parts you're buying and the boards you're re-working have lead-free pin finishes.
f) an assortment of tweezers and dental picks.
g) hot-air rework station with an assortment of nozzles.
h) a proper low-foaming cleaner like Detergent 8 (that's what it's called), diluted as necessary. Because with all of that flux, you'll have to clean the board.
i) stiff brush to clean the flux.
j) supply of deionized/distilled water for rinsing.
k) shop air to blow the boards dry after rinsing.
l) Solder paste, if you're adventurous.
m) those little quarter-size things of tip tinner are nice.

Soldering QFPs and SOICs is not that tough. Align the package. Tack-solder the corner pins. Apply flux, then "drag solder" all of the pins. You'll make a mess and the solder will blob over more than one pin. Get your thin solder wick and clean it up.

Soldering passives is likewise easy. Do not tin the pad, but do put down some flux. This prevents tombstoning.  Get some solder on the iron tip. Hold the part down with your tweezers while you touch the tip to the pad and to the end of the part. Let the solder flow. Do the other side of the part.

Need to remove an SMD chip or passive? Get out the hot-air tool. Select the right nozzle. Use the lowest velocity air and let it get hot. The idea is to get the air hot enough to melt the solder (which happens at what, 200C?) while not scorching the board and lifting up pads. Let the tool heat up the work while you poke underneath the chip with a dental pick or tweezers. When it's right, the solder will suddenly flow and the part will flow up, so flip it up/grab it with the tweezers and remove the air. Then grab your soldering iron, solder wick and some flux and clean up the pads.

Most people don't know this, but MLC ceramic capacitors were not designed for hand soldering. They're actually pretty easy to break. The big/wide parts (those great 10 uF parts used in switch-mode power supplies) are the worst for this. They tend to break if you have some solder on the pad before you put the part down and you then press on the part while soldering.

After soldering, clean the board. Spritz the detergent onto the board and scrub the flux off with your brush. Hint: a narrow painting "chip" brush, with all but about 3/16" of the bristles cut off, is a fabulous PCB scrub tool. Then rinse with the deionized water. Shop air from the little pistol nozzle at about 30 psi is great for blowing out the excess water. The detergent and using DI water make this easy, because the water tends to collect like water on a newly waxed car, and it just blows out.

Obviously, unsealed pots and switches aren't washable, so put them on last after assembly. If you're doing rework, then do your best to keep the cleaning agents and flux out of them.

Don't be scared of SMT. Impress your friends with your mad rework skills!

-a

"On the Internet, nobody can hear you mix a band."

Quote    Reply   
avatar

dcollins

Platinum Blonde

Posts: 2,326 Member Since:27/01/2011

#3 [url]

Apr 7 12 1:27 PM

Don't be scared of SMT. Impress your friends with your mad rework skills!

-andy_peters

Most excellent Andy, many thanks for that.

The thread title was really in jest, I'm not really a'feared of SMD and have done a fair amount already. No major damage done, although spending so much time looking through the stereo microscope gets pretty fatiguing.  Do you find this to be the case?

Fortunately, I'm dealing with complete boards from the prototype house and only tweaking a few parts, not removing the 144 pin package.  Until I blow one up, that is............

I did not know that about the fragility of MLCC caps.  Save for a handful of elcaps, the majority of the parts I used are Kemet C0G's and I'l be extra careful.

Thanks again, and I'm sure I'll have more questions.


DC

 davecollinsmastering.com


Quote    Reply   
avatar

andy peters

Silverado

Posts: 54 Member Since:07/02/2011

#5 [url]

Apr 7 12 5:05 PM

Most excellent Andy, many thanks for that.The thread title was really in jest, I'm not really a'feared of SMD and have done a fair amount already. No major damage done, although spending so much time looking through the stereo microscope gets pretty fatiguing.  Do you find this to be the case?

-dcollins

It's not so much the staring through the microscope that gets me. It's the sitting position and basically not moving that sends me to the chiropractor (who says, "Don't do that"). I do not know how the assembly techs can sit at a station for eight hours and not go nuts.

I did not know that about the fragility of MLCC caps.  Save for a handful of elcaps, the majority of the parts I used are Kemet C0G's and I'l be extra careful.

We (day job) didn't know this, either, until we started having issues with capacitor failures. In a lot of cases, you won't even know that the caps are bad. An example would be the ubiquitous 0.1uF bypass caps. There are so many caps on a board that if one was bad you'd never know. I think it was first noticed when during board test, a tech noticed that system noise was higher than it should have been, and it was traced to an oscillating regulator. The cap hanging on the regulator output was the proper value and type, but it was removed from the board (hot-air tool so as to not wreck it) and it was measured and it was basically a non-capacitor. Parts in the bin were tested, and they were fine, but that particular failure was noted across a handful of boards.

One of the techs spent some time looking through the Panasonic and the Kemet and the AVX web sites about ceramic SMT caps, and the manufacturers all pretty much say, "Don't hand-solder these caps." Then the docs basically say, "But since you're going to do it anyway, here are the recommendations ..." I'll dig them up on Monday for you.

Those docs made another point which we had not considered, which was that when the boards are machine-stuffed and the panels are broken apart after assembly, a part near where the shearing blade does its thing might be stressed by the board flexing. So there are recommendations about keeping parts some distance from the edge of a PCB, and aligning the parts in a certain direction, and some other little non-obvious design rules.

"On the Internet, nobody can hear you mix a band."

Quote    Reply   
avatar

cwhumphrey

Gold Finger

Posts: 343 Member Since:27/01/2011

#6 [url]

Apr 7 12 6:18 PM

Ha!

I used to have perfect vision. Now I have an astigmatism in my left eye. Sitting in "soldering position" for extended periods of time hurts my back.

I'm so glad I don't do this kind of work for a living anymore.

Having said that, let me echo a few ideas (and throw in some of my own):

Use a "pencil" solder tip. The tip should be in great shape.

I like a solder vise that swivels, has a large base and has slots in the jaws to hold a card.

You want a lot of light!

Use the right gauge of solder and solder wick for the job.

Use leaded solder until The Man takes it away. If you're using lead free solder, make sure you have an iron with enough wattage and ween yourself off of cleaning the tip with a wet sponge - it kills your heat and reduces the life of the tip.

Get the long needle nose pliers. They're weird to work with at first, but you'll be thanking me later.

Use a "flush cut" set of nippers. They're more expense, but again, you'll be thanking me later.

Use a flux brush to clean flux residual and don't use cleaner with any kind of lube in it.

Cheers,

Carter William Humphrey The WHITEST name in show business!™

Quote    Reply   
avatar

andy peters

Silverado

Posts: 54 Member Since:07/02/2011

#7 [url]

Apr 7 12 11:33 PM

 Use leaded solder until The Man takes it away. If you're using lead free solder, make sure you have an iron with enough wattage and ween yourself off of cleaning the tip with a wet sponge - it kills your heat and reduces the life of the tip.

-cwhumphrey

A 150W Weller "gun" will go cold when you try to solder to a plane or to heavy cable. It is NOT the wattage. It is the ability of the tool to maintain tip temperature when the load changes.

If the tip gets cold and then has to heat up again, you will NOT get a good solder joint.

If the tip maintains its temperature, you can use a 500-degree tip with lead-free solder and your work will be fine.

And yes, I clean the tip of the Metcal with a wet sponge.

Please, everyone, stop talking about how much power your soldering tool wastes. It's like saying how much power your speakers put out.

-a

"On the Internet, nobody can hear you mix a band."

Quote    Reply   
avatar

johnr

Platinum Blonde

Posts: 1,561 Member Since:25/01/2011

#8 [url]

Apr 8 12 5:28 AM

Those solder gun tips don't have enough heat storage capacity to not cool down when put in contact with a large piece of copper, which might be bigger than the tip itself if it's a heavy cable. A large poker sized soldering iron can be quite effective for heavy cable and large PCB planes (but useless for finer work obviously).

Quote    Reply   
avatar

johnr

Platinum Blonde

Posts: 1,561 Member Since:25/01/2011

#9 [url]

Apr 8 12 5:37 AM

Those docs made another point which we had not considered, which was that when the boards are machine-stuffed and the panels are broken apart after assembly, a part near where the shearing blade does its thing might be stressed by the board flexing.

-andy_peters

Another thing to watch out for is that these caps sometimes fail short circuit when they break. I usually use fail-open types for power supply decoupling.

Quote    Reply   
avatar

zenmastering

Silverado

Posts: 154 Member Since:21/02/2011

#10 [url]

Apr 9 12 2:03 AM

Thanks much for this very timely thread! Andy, your info is most appreciated, as I'm going to be doing some DAC/clock work in the near future and my SMD chops need some 'refreshing'

Whichever year it was (1984?) the Roland SRV2000 reverb came out, I was a tech at Roland and had to manually rework some of the DSP cards in that thing... a trial by fire into the world of SMD!

Cheers,

Graemme

Graemme Brown **Zen Mastering** Gabriola, Canada 1.604.874.9096 info AT zenmastering DOT net

Quote    Reply   
avatar

ssltech

Aqua Marine

Posts: 4,034 Member Since:22/01/2011

#11 [url]

Apr 9 12 7:29 AM

This is getting stickied... Plenty of great stuff!- Many thanks to Andy for the rich post, and to Dave for raising the subject!

-Keith Andrews -If I can't fix it, I can fix it so [i]NOBODY[/i] can fix it!

Quote    Reply   
avatar

dcollins

Platinum Blonde

Posts: 2,326 Member Since:27/01/2011

#12 [url]

Apr 9 12 5:24 PM

This is getting stickied... Plenty of great stuff!- Many thanks to Andy for the rich post, and to Dave for raising the subject!

-ssltech

The next topic will be passive components for audio that are available in SMD..........................


DC

 davecollinsmastering.com


Quote    Reply   
avatar

collins

Silverado

Posts: 51 Member Since:15/02/2011

#16 [url]

Apr 11 12 8:06 AM


And those magnifying visors make you look so cool.

-compasspnt



Geek chic...

[image]
Cheers,
Tim



hah...absolutely geek chic...
no kidding though, i sometimes use these all at the same time for ultimate coolness.  
Talk about needing a third/forth hand.


And yes, using these and/or a stereoscope for any time at all is VERY fatiguing.
I can only manage an hour or so at a time before i need to get up and look
at something far away to reset my eyes and neck.


Click here to view the attachment

Quote    Reply   
avatar

andy peters

Silverado

Posts: 54 Member Since:07/02/2011

#17 [url]

Apr 11 12 5:07 PM

I dug up some of the references I mentioned.

http://industrial.panasonic.com/ww/i_e/21088/smd-film-capacitor_e/smd-film-capacitor_e/soldering/soldering_failureeng.pdf

http://tdk.com/cap_technotes.php

http://www.ami.ac.uk/courses/topics/0179_fmcc/index.html

http://www.dfrsolutions.com/uploads/publications/2006_Cracking_Pb-free.pdf

http://industrial.panasonic.com/www-data/pdf/ABJ0000/ABJ0000PE1.pdf

Lots of data out there, but it's mostly targeted towards the folks who are in charge of large-volume production and not so much towards hobbyists and the DIY crowd. 

I suppose I should provide a bit of background. I'm an EE with about 24 years of working as a design guy, mostly now with FPGAs and microcontrollers. The company I work for now builds high-end scientific CCD cameras, and these products present some interesting challenges, not the least of which is that we typically cool the sensors down to -50C and below (dark current noise goes down as temperature goes down). Of course you can't just cool something to those temperatures in atmosphere, so the sensors and some of the electronics are mounted in vacuum chambers. And being in a vacuum presents all sorts of challenges regarding materials and cooling and such.

So I approach all of this in terms of best practices. It might seem like I'm shilling for Metcal, but I've used all of the Hakkos and Wellers that people like, and there's no comparison. And the truth is, if you want to play this game, you need the proper tools. Oscilloscopes and other test equipment can be expensive, but you're paying for reliability and quality of the measurement. My comparison is simple: you don't want to use a Mackie mixer or a Behringer microphone, so why should you use the equivalent in your shop tools? Yeah, the Mackie might sound good, until you hear the Neve.

Oh, and I used to be a full-time club sound guy (while working a day job as an EE) and I still work with a band from my home state of NJ. (I'll be in Brooklyn and Hudson, NY later this month, and Boston and Northampton, MA next month.) And I like to see what's going on inside of audio gear.

-a

"On the Internet, nobody can hear you mix a band."

Quote    Reply   
Add Reply

Quick Reply

bbcode help